PROJECT 4 ? SUMMARY ENVIRONMENT-GENOME INTERACTIONS The genome and molecular circuitry it encodes do not act alone, but are embedded in a web of interactions with environmental factors and stimuli such as nutrients, hormones, drugs, toxins and other chemical compounds. In this project, we will apply a multifaceted approach to systematically study the interactions between the organism and the environment, with the aim of developing large maps of gene-environment interactions and analyzing these maps to exploit fundamental principles and rules of interaction that can be used predictively. We will use two complementary network mapping approaches, one based on sequencing of resistant isolates and another based on a high-density chemogenetic screening platform. The project has three Specific Aims. Aim 1 is based on the principle that patterns of protective mutations acquired within a genome, when the genome is exposed to a toxic small molecule growth inhibitor, can be used to simultaneously explore protein-protein interactions and study how the organism's genome interacts with this environment. Instead of using a system of knockout strains that represent the entire genome, we will allow the genome to adaptively change when exposed to a small molecule and then analyze it systematically. A total of 60 drugs will be tested in yeast in this assay. These studies will then be mirrored in a haploid human cell line, to perform small-scale whole genome sequencing studies of how human cells mutate away from cancer drugs. Aim 2 will employ a complementary genetic mapping approach to further develop networks of genes that provide resistance to environmental perturbagens. Taking advantage of a new high-throughput 6144 colony screening format, we will screen a library of 125 chemical compounds in haploid S. cerevisiae strains that over-express or under-express complete genes, modeling the structural changes that develop in genomes exposed to environmental toxins. We will then test a subset of this interaction space among orthologous genes in human cancer cells. In the third aim, we propose to integrate the chemical-genetic interaction maps derived from evolutionary resistance (Aim 1) and genome-wide libraries (Aim 2) with a vast assortment of prior knowledge to build a cross-species model of drug-gene interaction. The goal of this model will be to learn how to predict human drug-gene interactions from diverse information including maps of drug-gene interactions in another species. Collectively, these aims will significantly advance our knowledge of the global gene networks that govern how cells become resistant to drugs. These efforts are led jointly by two pioneers in the systems biology of infectious disease: Dr. Elizabeth Winzeler, who formally joins the SDCSB faculty as of this renewal, and Dr. Sumit Chanda, a current SDCSB investigator. The project leverages the yeast genetic screening platform and computational analysis expertise established by Dr. Trey Ideker, as well as involvement from two SDCSB junior investigators, Dr. Hannah Carter and Dr. Jason Kreisberg.